ABSTRACT Breast cancer mortality is largely attributable to systemic, hematogenously-disseminated metastatic disease. Given the limitations of current prognostic criteria, new methods to identify tumors likely to metastasize and respond to therapy are needed. Multiphoton-based intravital imaging has shown that invasive carcinoma cells in rodent mammary tumors intravasate via peri-vascular structures containing a Mena over-expressing tumor cell, a macrophage, and an endothelial cell, all contacting each other. This tri-partite arrangement of cells facilitates entry of carcinoma cells into the blood vessel. We have identified this microenvironment in human breast cancer samples using a triple immunostain for formalin-fixed paraffin-embedded tissue that simultan- eously labels the 3 cell types. We call the direct apposition of these 3 cell types ?TMEM?, for Tumor Micro- Environment of Metastasis. Recently, in a cohort of patients at Kaiser Permanente (KP), we showed that TMEM was positively associated with risk of distant metastasis in ER+/HER2- breast cancer independently of IHC4, a composite immunohistochemical score (based on ER, PR, HER2, Ki67) that provides prognostic information comparable to the Oncotype Dx Recurrence Score (RS). Extension of this work to encompass comparison to other recently developed prognostic markers (e.g., PAM50) is now warranted. For TMEM, the invasive carcinoma cells are identified using a protein marker for invasive and migratory cancer cells called ?Mena?, which has multiple splice variants: Mena11a is an anti-metastatic isoform expressed in epithelial-like but not mesenchymal-like tumor cells, while Mena invasive (MenaINV) confers a potent pro-metastatic pheno- type in mesenchymal-like tumor cells. Recently, we used multiplex quantitative immunofluorescence to estimate the abundance of Mena lacking its anti-metastatic Mena11a isoform. We showed that this marker, Menacalc, which reflects the relative amount of epithelial-mesenchymal transition (EMT) that a tumor has undergone so that tumor cells can participate in TMEM assembly and interact with TMEMs, was positively associated with risk of breast cancer death. We now propose to: examine the association of these markers (Menacalc/MenaINV/TMEM) with risk of distant metastasis both in a case-control study of 600 case-control pairs nested in an expanded KP cohort of 8769 breast cancer cases, and in 1000 breast cancer cases sampled from the B28 trial (which assessed the value of paclitaxel as adjuvant chemotherapy); examine the association with risk of distant metastasis of the strongest TMEM-related marker or marker combination in comparison to and in addition to IHC4 and PAM50 (KP study) and Oncotype Dx RS and PAM50 (B28 study) [~2/3 of the B28 study sample will have Oncotype Dx RS results available]); examine whether TMEM-related marker or marker combination predicts response to therapy (in B28); and externally validate in the B28 study population the TMEM/Mena score (the combination of TMEM, Menacalc, MenaINV most strongly associated with risk) developed in the KP study population.